[ad_1]
In Word: Although global economies have started to open up, the Covid-19 virus is still spreading around the world, infecting thousands of people every day. To help slow the spread of the disease, MIT researchers have developed an AI model that can detect the presence of the virus even in asymptomatic people.
The potential good that such a model could do is probably pretty obvious. Suppose the model was refined and released to the general public in some way, perhaps in the form of a free mobile app.
In that case, it could help people detect the infection and get tested or avoid contact with others, if necessary.
Teachers, for example, might use it every day before going to class, just like other people who work around strangers; Frontline retail employees are another key audience for such a tool.
Fortunately, migrating the model to an application is precisely what the researchers are working on now. However, the team has yet to finish developing it, and of course get FDA approval before it can be widely distributed.
However, the first results are promising. To date, the model has been trained on “tens of thousands” of cough samples as well as “spoken words.” When the researchers input new cough recordings into the model, they can accurately identify full-fledged Covid-19 infections in “98.5 percent of coughs” and in 100 percent of asymptomatic people.
It seems that even for people who do not show severe symptoms of Covid-19, the way they cough, breathe and speak can contain small indicators that can point towards a possible infection.
If this model is really as accurate as researchers believe, we hope the FDA will approve it sooner rather than later; the world could certainly use it right now.
